Federalism: Evaluating the Founders' Design

University of Chicago Law School Chicago Unbound

Journal Articles Faculty Scholarship

1987

Federalism: Evaluating The Founders' Design

Michael W. McConnell

Follow this and additional works at: https://chicagounbound.uchicago.edu/journal_articles

Part of the Law Commons

Recommended Citation Michael W. McConnell, "Federalism: Evaluating The Founders' Design," 54 University of Chicago Law Review 1484 (1987).

This Article is brought to you for free and open access by the Faculty Scholarship at Chicago Unbound. It has been accepted for inclusion in Journal Articles by an authorized administrator of Chicago Unbound. For more information, please contact [email protected]. REVIEWS

Federalism: Evaluating the Founders'* Design Michael W. McConnell Federalism: The Founders' Design. Raoul Berger. University of Oklahoma Press, 1987. Pp. 223.

Raoul Berger stands for the honorable tradition that a scholar must put aside his own social and economic predilections and look only to original sources in seeking the meaning of the United States Constitution. His numerous books and articles have ad- dressed many of the most pressing questions about the organization and prerogatives of the three branches of the federal government, including the President's right to withhold information from Congress,1 the congressional power of impeachment,2 the reach of executive war powers,3 the scope of judicial authority under the 5 fourteenth amendment,4 the legitimacy of capital punishment, and the congressional power to remove controversial matters from the appellate jurisdiction of the Supreme Court.6 On some of these issues (executive privilege, impeachment, war powers, and jurisdic-

* 0 1987 by Michael W. McConnell. t Assistant Professor of Law, The University of Chicago. Thanks are due the Russell Baker Scholars Fund for providing financial support during the preparation of this review, and to Albert Alschuler, Mary Becker, David Currie, Frank Easterbrook, Richard Helmholz, Dennis Hutchinson, Fred McChesny, Geoffrey Miller, Richard Posner, Geoffrey Stone, and Cass Sunstein for helpful comments on an earlier draft. [Editors' note: This issue contains two contrasting reviews of Raoul Berger's Federal- ism: The Founders' Design. Although Professors McConnell and Powell argue on behalf of significantly different positions, their articles are not in the form of a "debate." Neither reviewer had an opportunity to see the other's manuscript prior to publication of the issue.] I Raoul Berger, Executive Privilege: A Constitutional Myth (1974). 2 Raoul Berger, Impeachment: The Constitutional Problems (1973). 3 Berger, Executive Privilege at 60-116 (cited in note 1). 4 Raoul Berger, Government By Judiciary- The Transformation of the Fourteenth Amendment (1977). s Raoul Berger, Death Penalties: The Supreme Court's Obstacle Course (1982). Raoul Berger, Congress v. The Supreme Court (1969).

1484 Evaluating the Founders' Design 1485 tion stripping) Berger has taken the position usually associated with political liberals. On some (capital punishment and the fourteenth amendment) he has taken a position more closely associated with political conservatives-indeed, in the case of the fourteenth amendment going far beyond them. On some he has taken positions explicitly contrary to his own political preferences.7 In his new book Federalism:The Founders' Design, Berger for the first time addresses the Constitution's allocation of power between the federal government and the states. Those who have ears to hear will find evidence more than sufficient to show that what the people ratified is something quite different from what they ul- timately got. The book is neither analytical nor theoretical; the reader must bring to the book his own framework for translating historical detail into a "usable past."8 The strength of the book is entirely in its details-in the relentless collation of quotations in the text, with still more piled in the footnotes, demonstrating, in a variety of contexts, that the framers and ratifiers of the Constitu- tion intended the authority of the states to be far greater, and that of the federal government far less, than it has turned out to be.' Berger's work casts doubt on (he would say clearly refutes) most of the major arguments justifying today's centralization of authority in Washington. This is not to say, however, that The Founders' Design is a particularly successful book. Even to one sympathetic, as I am, to Berger's position, the book seems radically incomplete. While it provides ample reason to believe our constitutional structure has gone astray, it gives no reason why we should care and no attention to the considerations of democratic authority and judicial restraint that have contributed to the present state of affairs.

I. The Founders' Design shares the weakness of Berger's earlier writing, in that it fails to link his discoveries about the issue at

See, e.g., Berger, Government By Judiciary at 4 (cited in note 4). Indeed, Professor Berger announces that his conclusions in The Founders'Design "not infrequently are at war with my predilections" (p. 5). All parenthetical page references in text and notes are to Raoul Berger, Federalism: The Founders' Design (1987). 8 Akhil Reed Amar, Of Sovereignty and Federalism, 96 Yale L.J. 1425, 1426 n.9 (1987)(emphasis in original). 9 The book could have used more careful editing. Quotations are often repeated twice and more, for no apparent reason. In the most irritating instance, Berger quotes Chief Jus- tice Marshall in a footnote, and then uses the same quotation in the succeeding sentence in text. (pp. 155-56 and n.162). 1486 The University of Chicago Law Review [54:1484 hand to any overarching understanding of the purposes and archi- tecture of the Constitution. Instead, it is narrowly and dispiritingly positivistic. Great constitutional scholarship is, like Berger's, atten- tive to the details of the document and true to its sources. But it also does something more (and this something is what makes constitutional law a worthwhile scholarly enterprise): it makes the Constitution a window through which we learn about humankind as a political creature. The United States Constitution inspires reverence not just because it was drafted and ratified by our forefathers, who were an uncommonly clever lot, but because it is the most successful attempt in history to construct a polity consistent with both the baser passions and the higher aspirations of its citizens. Studying the Constitution has some of the same intellectual delight as reading Aristotle: it opens the mind on a subject of the first importance. The Founders' Design has little of this quality. We learn from it that our forefathers had a high regard for the autonomous exis- tence of the states and intended the Constitution to guarantee this autonomy. We do not learn how this scheme of dual sovereignty relates to any of the great themes and objectives of American constitutionalism. We are, therefore, shut off from an important source of wisdom about things political and left indifferent, except as a technical matter of fealty to positive law, to the identified de- partures from the founders' design. The Founders' Design also exhibits a confusion, less evident in Berger's earlier books, about the theoretical underpinning of his constitutionalism. This book is not, like Government By Judici- ary'° or Death Penalties: The Supreme Court's Obstacle Course," a call for judicial restraint. It gains no support from a theory of constitutional interpretation based on democratic majoritarianism. If carried into practice, The Founders' Design would require wholesale, and not always predictable, judicial intervention into American governance and a massive repudiation of laws duly adopted by representative bodies and supported by large majorities of the populace. It therefore presents a break with Berger's prior posture of virtual legislative supremacy. 2 The usual position of those who rail against "government by judiciary" is that democratic decision making is the general rule,

10 Berger, Government by Judiciary (cited in note 4). 11Berger, Death Penalties (cited in note 5). 11See, e.g., Berger, Executive Privilege at 3 (cited in note 1). See also The Founders' Design (p. 14) (calling the legislative branch the "darling of the Founders"). 1987] Evaluating the Founders' Design 1487 with judicial intervention justified only when the Constitution fairly can be interpreted (in light of its text, structure, history, and purposes) as foreclosing the course of action adopted by representative institutions. Because decisions holding governmental choices unconstitutional cannot be reversed, short of constitutional amendment or change of Court personnel, a court should resolve doubts against intervention. This is the lesson of Dred Scott,1" which stripped Congress of its power to bring slavery to heel, and also of the New Deal, which had to confront judge-made barriers to national social and economic legislation. Of the two classes of judicial error-striking down constitutional legislation and upholding unconstitutional legislation-the former is the more dangerous, since the political corrective is so much more difficult. In The Founders'Design, however, Berger refuses to recognize that most of the Supreme Court's retreat from federalism has been a product of deference to democratic choice and to the tendency to avoid the more dangerous type of error, even at the cost of the less dangerous one. Indeed, he fails to acknowledge the seemingly obvious conflict between judicial restraint and the aggressive judicial enforcement of federalism principles. "It is to be hoped," Berger comments in the book's concluding chapter, "that the historical facts may lead the court to curtail its increasing intrusion into the States' internal affairs" (pp. 187-88) (emphasis added). He qualifies his hope by adding that "those who enjoy the exercise of uncurbed power are unlikely to surrender it merely because it has been usurped" (p. 186). These comments might be closer to the mark if they dealt with Roe v. Wade,14 or with federal court takeover of state prisons and hospitals, or with judicial supervision of state codes of criminal procedure. In these contexts the courts might be said to have "usurped" decisionmaking authority constitutionally vested in the state governments. But how can Berger say this of decisions upholding Acts of Congress, duly passed by representatives of the people? Is there no difference between judicial aggran- dizement and judicial deference to legislative authority? Berger needs to develop a new, more comprehensive theory of constitutional interpretation. His prior anti-judicial, pro-legislative stance cannot explain his conclusions here. I do not mean to suggest that Berger's position cannot be defended; a convincing case can be made for a more active judicial role where the courts deter- mine the allocation of decision making authority among represen-

,S 60 U.S. (19 How.) 393 (1857). 410 U.S. 113 (1973). 1488 The University of Chicago Law Review [54:1484

tative institutions rather than arrogating final decision making power to themselves.15 Without such a theory, however, the mes- sage of The Founders' Design is less than compelling. In any event, Berger's contrast between the original constitutional design and the current situation is somewhat overdrawn. The erosion of local autonomy may well have been inevitable, given the constitutional structure. Whatever the founders' intentions, the rules they wrote are skewed in favor of national power. In cases of conflict between state and federal law, federal law wins.16 If there is disagreement over constitutional rules, a depart- 17 ment of the federal government, the courts, serves as umpire. And the principal structural protection for federalism, the direct representation of state legislatures in the Senate, was eliminated by the seventeenth amendment.18 Technological and social change also play a part. Constitu- tional limits expressed in terms of interstate consequences lead to different results when applied to railroads than when applied to a horse and buggy. As the size of the market has expanded, so has federal power. Furthermore, for most people most of the time, issues of federalism take second seat to particular substantive outcomes. Even a conservative administration, ostensibly committed to a restoration of federalism, backed legislation coercing the states to raise their drinking age, a matter reserved to the states under the twenty-first amendment.19 The measure was then upheld in an opinion by the states' supposed best friend, Chief Justice Rehn- quist.20 So it has gone for 200 years. Berger does not comment on these aspects of the "founders' design." Instead, he places the en- tire blame on the courts.

"6This approach is diametrically opposed to the one advocated by Professor Herbert Wechsler and Dean Jesse Choper. See Herbert Wechsler, The Political Safeguards of Feder- alism: The Role of the States in the Composition and Selection of the National Government, 54 Colum.L.Rev. 543 (1954); Jesse H. Choper, Judicial Review and the National Political Process (1980). 16 U.S.Const. art. VI, § 2 (supremacy clause). 1" Marbury v. Madison, 5 U.S. (1 Cranch) 137 (1803). But see Garcia v. San Antonio Metropolitan Transit Authority, 469 U.S. 528, 547-554 (1985) (leaving state constitutional claims to the mercies of Congress). " U.S.Const. amend. XVII ("The Senate of the United States shall be composed of two Senators from each state, elected by the people thereof, for six years .... "). 19 See 324 Liquor Corp. v. Duffy, 107 S.Ct. 720, 726-27 (1987); id. at 730-34 (O'Connor dissenting). 20 South Dakota v. Dole, 107 S.Ct. 2793 (1987). 1987] Evaluating the Founders' Design 1489

II.

The Founders' Design begins (p.21) with the question: which came first, the nation or the states? After the Revolution and before ratification of the Constitution, were the states sovereign governments, or did their authority devolve from the Union? Ber- ger persuasively argues, contrary to the positions of Joseph Story and of Justice George Sutherland in the Curtiss-Wright case,2 1 that after the Declaration of Independence, each of the colonies became fully independent states-independent of each other as well as of the mother country. Indeed, a literal reading of the Arti- cles of Confederation, eventually adopted by all thirteen states, can hardly yield any other conclusion. Berger is less persuasive when explaining why this matters. The important question is not the locus of sovereignty prior to the Constitution, but under the Constitution. Berger's presentation implies a continuity in the theory of sovereignty from 1776 to 1787; but he undertakes no explicit defense of this assumption. He ignores the Constitution's own evidence on the issue: that it conspic- uously drops the Articles of Confederation provision stating that "each state retains its sovereignty, freedom and independence;" 22 that it boldly states the source of its authority as "We, the People of the United States; '23 that for ratification it bypassed the state legislatures and went to the people directly through conventions. The most persuasive inference from the text of the Constitution is that sovereignty rests in the people of the United States, and not in the governments of either the states or the nation.25 It follows that the extent of federal authority is neither more nor less than that delegated through the Constitution. That the states were in some sense sovereign prior to the Constitution seems largely irrelevant. The initial discussion of sovereignty reveals a tendency,

2' Joseph Story, 1 Commentaries on the Constitution of the United States §§ 207-217 at 142-54 (2d ed. 1851); United States v. Curtiss-Wright Export Corp., 299 U.S. 304, 316-19 (1936). 2 U.S. Articles of Confederation art. II. 23 U.S.Const. preamble. Note that the term "United States," even under the Articles, referred to the confederacy rather than to the several states, and that every reference to the United States in the Constitution is to the federal government. See, e.g., U.S. Const. amend. X ("The powers not delegated to the United States by the Constitution. . . are reserved to the States respectively . . "). 21 U.S. Const. art. VII. See Federalist 22 (Hamilton), in Clinton Rossiter, ed., The Fed- eralist Papers 152 (1961). 25 For a more extended discussion, see Amar, 96 Yale L.J. at 1429-39 (cited in note 8). 1490 The University of Chicago Law Review [54:1484 present throughout The Founders' Design, to emphasize what the founders said about the Constitution rather than the words of the Constitution itself. While the founders' explanations of their purposes are obviously useful in helping us to understand the words of the Constitution, Berger teeters on the brink of saying that the meaning of the Constitution is subordinate to the founders' intentions. At one point he states, quoting from Hawaii v. Mankichi,26 "'The intention of the lawmaker is the law,' rising even above the text" (pp. 15-16). Elsewhere, he states that "judges are confined to the four corners of the Constitution as explained by the Founders" (p. 10)-a position that (assuming "the four corners of the Consti- tution" to include arguments based on structure, history, and purposes as well as bare text) rests more comfortably within the tradi- tional understanding of constitutionalism. These two statements-that intentions rise above the text and that judging is confined to the four corners of the document-are irreconcilable and suggest that Berger is in something of an interpretive muddle. This muddle manifests itself in two ways throughout The Founders' Design. First, it leads him to underplay arguments from the text of the Constitution. Berger is far too quick to give up on the words of the Constitution; indeed, I cannot recall a single argument in the book that depends, in a serious way, on textual analysis. Second, it leads him to confuse the founders' expectations about how the nation would be governed under the Constitution with the founders' understanding of the meaning of the Constitu- tion. This problem pervades the book. The succeeding chapters, on the tenth amendment, the general welfare clause, the commerce clause, the fourteenth amendment, and the specific issue of mass transit, each contrasts the founders' expectations about the polity they were constructing with the way things have turned out. But this is not necessarily the same question as whether modern developments have violated the Constitution. For example, did the founders expect agriculture to become an important element in national commercial life? I agree with Berger (pp. 73-75) that they did not. Hamilton, no advocate of "states' rights," wrote that "the supervision of agriculture and of other concerns of a similar nature . . . which are proper to be provided for by local legislation, can never be desirable cares of a general jurisdiction" (p.74).27 Does it follow that the Congress of 150 years

28 190 U.S. 197, 212 (1903). 17 Federalist 17 (Hamilton), in The Federalist Papers at 118 (cited in note 24). 1987] Evaluating the Founders' Design 1491 later acted illegitimately when it concluded that regulation of agriculture was a "necessary and proper" means for curing national economic depression? The framers and ratifiers of the Constitution established rules and standards for determining the scope of national authority; that those rules and standards produce different outcomes in later circumstances is neither surprising nor troubling. The legal question must be whether congressional agricultural regulation is sufficiently related to "Commerce. . .among the several States." On that issue, the founders' expectations about agriculture are interesting and important, but cannot take precedence over the constitutional standard.

III.

The biggest disappointment in The Founders'Design is that it ignores the intellectual case for federalism. That the states should retain substantial independent authority is not self-evident. As Madison pointedly inquired in Federalist 45:

[I]f. . .the Union be essential to the happiness of the people of America, is it not preposterous to urge as an objection... that such a government may derogate from the importance of the governments of the individual States? Was, then, the American Revolution effected, was the American Confederacy formed, was the precious blood of thousands spilt, and the hard-earned substance of millions lavished, not that the people of America should enjoy peace, liberty, and safety, but that the government of the individual States, that particular municipal establishments, might enjoy a certain extent of power and be arrayed with certain dignities and attributes of sovereignty?28 Madison's question was directed to "adversaries to the plan of the convention. ' ' 29 It might as aptly be directed to the adversaries of our present centralized government: why forego national measures, thought to promote the peace, liberty, and safety of the people, merely because they intrude upon the "certain extent of power" traditionally reserved to the governments of the individual States? Why do we care about federalism? The Founders' Design ought to be the place to find answers. After all, the defenders of decentralized government in 1787-89 de-

28 Federalist 45 (Madison), in The Federalist Papers at 288-89 (cited in note 24). 29Id. at 288. 1492 The University of Chicago Law Review [54:1484 feated the more expansive proposals of the nationalists, eventually converting even Madison himself to their cause. Much of this was accomplished through force of argument. Even as they failed to block the "plan of the convention," they forced important com- promises, including addition of the Bill of Rights, and in the process became full co-founders of our innovative federal system. One of the most important developments of modern intellectual history has been the rediscovery of "federalist" and "anti-federalist" political thought.30 This development has coincided with the "public choice" movement in economics and political science, which lays the theoretical groundwork for an appreciative appraisal of the founders' thought.31 We are, therefore, better prepared than ever before to deal with Madison's question. Surprisingly, however, The Founders' Design gives little or no attention to reasons offered by the decentralizers, and accepted by the nation, for preserving a large measure of local autonomy. As a consequence, Berger is unable to subject their ideas to critical analysis. While asserting that the attachment to local governance "had a rational basis" (p. 55), Berger gives little clue as to what it was. It seems almost enough that the people of 1787 had a "natural attachment, 3 2 in Madison's words, to their states (p.56). The founders' design is much richer than Berger's book reveals. During the debates over the drafting and ratification of the Constitution, supporters and opponents alike came to articulate complex and sophisticated theories of federalism (which, it should be stressed, was a uniquely American blend of national systems-like the French-and confederate systems-like the ancient Greek and early modern Dutch).33 The "natural attachment" of the people in 1787 to their states was augmented by practical arguments about how the new system of dual sovereignty would promote three complementary objectives: (1) "[t]o secure the public good," (2) to protect "private rights," and (3) "to preserve the spirit and form of popular government." Achievement of these

o To my mind, the most interesting and important work is Herbert J. Storing, ed., The Complete Anti-Federalist (1981), including Storing's introductory essay, "What the Anti. Federalists Were For." 1, See Vincent Ostrom, The Political Theory of a Compound Republic (2d ed. 1987); James M. Buchanan and Gordon Tullock, The Calculus of Consent (1962). 32 Federalist 46 (Madison), in The Federalist Papers at 294 (cited in note 24). 11 See Federalist 18-20 (Madison), in The Federalist Papers at 122-38 (cited in note 24); Alexis de Tocqueville, Democracy in America 156 (J.P. Mayer ed. 1969); Martin Diamond, What the Framers Meant By Federalism, in Ronald A. Goldwin, ed., A Nation of States 24 (1963). 1987] Evaluating the Founders' Design 1493 ends, according to Madison, was the "great object" of the Consti- tution.34 To understand the "founders' design" we must look again at those arguments. As the people of 1987, we must look at them in light of modern experience and knowledge about political decision making. The arguments of 1787 stand up remarkably well.

A. To "Secure the Public Good" Rejecting both pure confederation and consolidation, the "Federal Farmer" (the ablest and most influential of the anti-federalist pamphleteers) argued that a "partial consolidation" is the only system "that can secure the freedom and happiness of this people." He reasoned that "one government and general legislation alone, never can extend equal benefits to all parts of the United States: Different laws, customs, and opinions exist in the different states, which by a uniform system of laws would be unreasonably invaded. 3 5 The framers sought, the Federal Farmer concluded, to preserve decentralized decision making because smaller units of government are better able to further the interests and general welfare of the people. Three important advantages of decentralized decision making emerge from an examination of the founders' arguments and the modern literature. First, decentralized decision making is better able to reflect the diversity of interests and preferences of individuals in different parts of the nation. Second, allocation of decision making authority to a level of government no larger than necessary will prevent mutually disadvantageous attempts by communities to take advantage of their neighbors. And third, decentralization al- lows for innovation and competition in government. 1. Responsiveness to diverse interests and preferences. The first, and most axiomatic, advantage of decentralized government is that local laws can be adapted to local conditions and local tastes, while a national government must take a uniform-and hence less desirable-approach. So long as preferences for government policies are unevenly distributed among the various locali- ties, more people can be satisfied by decentralized decision making than by a single national authority. This was well understood by the founding generation. A noted pamphleteer, "The Impartial Ex- aminer," put the point this way: "For being different societies, though blended together in legislation, and having as different in-

" Federalist 10 (Madison), in The Federalist Papers at 80 (cited in note 24). 35 Storing, 2 Complete Anti-Federalist at 2.8.13-14 (cited in note 30). 1494 The University of Chicago Law Review [54:1484 terests; no uniform rule for the whole seems to be practicable."' 6I For example, assume that there are only two states, with equal populations of 100 each. Assume further that 70 percent of State A, and only 40 percent of State B, wish to outlaw smoking in public buildings. The others are opposed. If the decision is made on a national basis by a majority rule, 110 people will be pleased, and 90 displeased. If a separate decision is made by majorities in each state, 130 will be pleased, and only 70 displeased. The level of sat- isfaction will be still greater if some smokers in State A decide to move to State B, and some anti-smokers in State B decide to move to State A.3 7 In the absence of economies of scale in government services,38 significant externalities, 9 or compelling arguments from justice,40 this is a powerful reason to prefer decentralized government. States are preferable governing units to the federal government, and local government to states. Modern public choice theory provides strong support for the framers' insight on this point.4' 2. Destructive competition for the benefits of government. A second consideration in designing a federal structure is more equivocal. The unit of decision making must be large enough so that decisions reflect the full costs and benefits, but small enough that destructive competition for the benefits of central government

" Storing, 5 Complete Anti-Federalist at 5.14.6 (cited in note 30). See also Tocqueville, Democracy in America at 161 (cited in note 33) ("In large centralized nations the lawgiver is bound to give the laws a uniform character which does not fit the diversity of places and of mores."). 37 Under certain extreme assumptions, a sufficiently decentralized regime with full mobility could perfectly satisfy each person's preferences even with no voting at all. See Dennis C. Mueller, Public Choice 126-29 (1979); Charles M. Tiebout, A Pure Theory of Local Ex- penditures, 64 J.Pol.Econ. 416 (1956). This point is also pertinent to the "liberty" argument for federalism. See notes 80-81 and accompanying text. 38 Economies of scale are probably not a major consideration. Small units of government are able to contract with one another or with private service providers so as to achieve economies of scale without sacrificing decision making autonomy. Gordon Tullock, Federal- ism: Problems of Scale, 6 Pub.Choice 19, 21 (Spring 1969). For a contrasting view, see Je- rome Rothenberg, Local Decentralization and the Theory of Optimal Government, in Julius Margolis, ed., The Analysis of Public Output 31, 33 (1970). 39 See notes 42-58 and accompanying text. 40 See notes 93-95 and accompanying text. 41 See Wallace E. Oates, Fiscal Federalism 11-13, 54-63 (1972); Tullock, 6 Pub.Choice at 19 (cited in note 38); J. Roland Pennock, Federal and Unitary Government-Disharmony and Frustration, 4 Behavioral Science 147 (Apr. 1959). The model in the text is oversimpli- fied. As Susan Rose-Ackerman has shown, under a decentralized regime citizens of a given state might support a policy at the state level while opposing the same policy on a uniform basis at the national level. Her example is legalized gambling: citizens of a state with legalized gambling have more to gain if most other states do not have it. Susan Rose-Ackerman, Does Federalism Matter?, 89 J.Pol.Econ. 152, 154-57 (1981). This qualification does not af- fect my argument. 1987] Evaluating the Founders' Design 1495

action is minimized. In economic language, this is the problem of "externalities. ' '42 Externalities present the principal countervailing consideration in favor of centralized government: if the costs of government action are borne by the citizens of State C, but the benefits are shared by the citizens of States D, E, and F, State C will be unwill- ing to expend the level of resources commensurate with the full social benefit of the action.43 This was the argument in Federalist 25 for national control of defense. 44 Since an MX missile in Penn- sylvania will deter a Soviet attack on Connecticut and North Caro- lina as well as Pennsylvania, optimal levels of investment in MX's require national decisions and national taxes. Or, similarly, since expenditures on water pollution reduction in Kentucky will benefit riparians all the way to New Orleans, it makes sense to nationalize decisions about water pollution regulation and treatment. Thus, as James Wilson explained to the Pennsylvania ratifying convention, "[w]hatever the object of government extends, in its operation, beyond the bounds of a particular state, should be considered as belonging to the government of the United States" (quoted at pp. 169-70) (emphasis in original). That significant external effects of this sort provide justifica- tion for national decisions is well understood-hence federal fund- ing of defense, interstate highways, national parks, and medical re- search, and federal regulation of interstate commerce, pollution, and national labor markets. It is less well understood that nation- alizing decisions where the impact is predominantly local has an equal and opposite effect. The framers' awareness that ill consequences flow as much from excessive as from insufficient centralization is fundamental to their insistence on enumerating and thus limiting the powers of the federal government. Hence the other half of Wilson's explanation: "Whatever object of government is confined in its operation and effect, within the bounds of a particular State, should be considered as belonging to the government of 4 that State" (p. 169) (emphasis in original). ' This stands in marked

"ISee generally Tullock, Federalism: Problems of Scale, 6 Pub.Choice at 19 (cited in note 38); Richard A. Posner, Economic Analysis of Law 600 (3d ed. 1986). 13 See Mancur Olson, Jr., The Principle of "Fiscal Equivalence:" The Division of Re- sponsibilities Among Different Levels of Government, 59 Am.Econ.Rev. 479, 482 (1969). " Federalist 25 (Hamilton), in The Federalist Papers at 163 (cited in note 24). For a modern version of the argument, see Mancur Olson, Jr. and Richard Zeckhauser, An Eco- nomic Theory of Alliances, 48 Rev.Econ. & Stat. 266, 278 (1966). 4B Wilson's formulation was widely echoed in the debates of the period. See, e.g., Stor- ing, 3 Complete Anti-Federalist at 3.14.8 (cited in note 30) (essays of "A [Pennsylvania] 1496 The University of Chicago Law Review [54:1484 contrast to the modern tendency to resolve all doubts in favor of federal control. The point is quite general. It applies to lawmaking and regulation no less than to taxing and spending.46 A major effect of regulation is to shift burdens from one region or locality to another. Fa- miliar examples are environmental laws that protect eastern "dirty" coal from competition from western "clean" coal4 and rail- road regulation that enables low density areas to maintain service at the expense of other traffic. 48 But the effect is especially obvious in the case of federal spending. If the national treasury is seen as a common pool resource for financing schemes of predominantly local benefit, it will be oversubscribed. Current budgetary woes are '49 largely attributable to this fiscal "tragedy of the commons. Where the benefits of government action are predominantly local but financing is national, each community can be expected to pursue projects even where total cost exceeds the actual benefit. Local decisionmakers will take into account only the local portion of the cost, since the national portion will be effectively "free." Nobel laureate James Buchanan has demonstrated mathemat- ically that centralized decision making about projects of localized impact will result in excessive spending-excessive meaning more than any of the individual communities involved would freely choose.50 Each community would be better off if they could agree in advance (as they thought they did in the Constitution) to con- fine federal attention to issues of predominantly interstate consequence. In this connection, Berger's chapter on the spending power is especially weak. Article I, § 8, clause 1 of the Constitution grants Congress the power "[t]o lay and collect Taxes, Duties, Imposts and Excises, to pay the Debts and provide for the common Defense

Farmer"). According to Tocqueville, a similar model was employed for the allocation of power between states and townships. See Tocqueville, Democracy in America at 67 (cited in note 33)("In all that concerns themselves alone the townships remain independent bodies, and I do not think one could find a single inhabitant of New England who would recognize the right of the government of the state to control matters of purely municipal interest."). '" See Storing, 5 Complete Anti-Federalist at 5.14.6 (cited in note 30)(the "Impartial Examiner" arguing that a single national mode of taxation will result in each state endeav- oring "to raise a revenue by such means, as may appear least injurious to its own interest"). 47Bruce A. Ackerman and William T. Hassler, Clean Coal/Dirty Air (1981). 48 See Southern Railway v. North Carolina, 376 U.S. 93 (1964). 49 See Garrett Hardin, The Tragedy of the Commons, 162 Science 1243 (1968). " Buchanan and Tullock, Calculus of Consent at 135-40 (cited in note 31). See also Olson, "Fiscal Equivalence" at 482-83 (cited in 43); Gordon Tullock, Comment, in Julius Margolis, ed., The Analysis of Public Output 65 (1970). 1987] Evaluating the Founders' Design 1497 and general Welfare of the United States." Berger goes to some lengths to support Madison's argument that this is not "an unlimited commission to exercise every power which may be alleged to be necessary for the common defense or general welfare" (p.102)." This is uncontroversial: in modern constitutional litigation no one cites the general welfare clause in support of federal power other than taxing or spending.5 2 The controverted issue is whether the clause places any limit on the objects for which Congress may tax and spend. On this point, Berger defends Madison's argument that spending for the "general welfare" is confined to spending for purposes elsewhere enumerated in the Constitution (p. 104). I am inclined, like Alexander Hamilton, Joseph Story, and the Supreme Court, to reject this argument on textual grounds. 3 "General Welfare" is a broader term than "purposes hereinafter enumerated," and Madison's interpretation makes the general welfare clause redun- dant. Yet Madison's argument is perfectly respectable, if only because Madison made it. However, Berger's presentation of the argument, especially his attempted refutation of Hamilton, is unsound. Essentially, Berger argues that because Hamilton unsuc- cessfully argued in favor of "the unlimited power of passing all laws without exception, ' 54 an unlimited power to spend money for the general welfare must have been equally "unacceptable to the Convention" (p. 108). This is to say that rejection of a broader power implies rejection of a narrower. In his rush to discredit Hamilton's position, Berger fails to note that Hamilton himself articulated a limitation on the spending power more theoretically satisfactory and more textually sup- portable than Madison's. "[T]he object to which an appropriation of money is . . .made [must] be General and not local; its operation extending in fact, or by possibility, throughout the Union, and not being confined to a particular spot."55 This construction is a

" Federalist 41 (Madison), in The Federalist Papers at 262 (cited in note 24). 52 See Story, 1 Commentaries on the Constitution § 907 at 630 (cited in note 21); Ed- ward Corwin, The Spending Power of Congress-Apropos the Maternity Act, 36 Harv.L.Rev. 548, 551 (1923). " United States v. Butler, 297 U.S. 1, 64-66 (1936); Alexander Hamilton, Report on Manufactures, in Harold L. Syrett, ed., 10 The Papers of Alexander Hamilton 302-04 (1966); Story, 1 Commentaries on the Constitution § 14 at 629 (cited in note 21). But see David P. Currie, The Constitution in the Supreme Court: The New Deal, 1931-1940, 54 U.Chi.L.Rev. 504, 534-36 (1987). " Max Farrand, ed., 1 The Records of the Federal Convention of 1787 300 (1937)(emphasis in original). 5' 10 Papers of Hamilton at 303 (cited in note 53)(emphasis in original). 1498 The University of Chicago Law Review [54:1484 persuasive reading of the term "general welfare" (the word "general" is frequently used in the debates to signify "national"56), and it guards against precisely the fiscal tragedy of the commons discussed above. Moreover-though Berger fails to mention it-early debates in Congress on a proposal to provide $15,000 for the relief of survivors of a fire in Savannah, Georgia support Hamilton's view that the line was drawn between objects of a predominantly local, as opposed to a general or national, impact.5" 3. Innovation and competition in government. A final reason why federalism has been thought to advance the public good is that state and local governmental units will have greater opportunity and incentive to pioneer useful changes. A consolidated national government has all the drawbacks of a monopoly: it stifles choice and lacks the goad of competition. Lower levels of government are more likely to depart from established consensus simply because they are smaller and more numerous. Elementary statistical theory holds that a greater number of independent observations will produce more instances of devia- tion from the mean. If innovation is desirable, it follows that decentralization is desirable. This statistical proposition is strength- ened, moreover, by the political reality that a smaller unit of government is more likely to have a population with preferences that depart from the majority's. It is, therefore, more likely to try an approach that could not command a national majority.58 Perhaps more important is that smaller units of government have an incentive, beyond the mere political process, to adopt popular policies. If a community can attract additional taxpayers, each citizen's share of the overhead costs of government is proportionately reduced. Since people are better able to move among states or communities than to emigrate from the United States, competition among governments for taxpayers will be far stronger at the state and local than at the federal level. Since most people are tax-

16 See, e.g., Storing, 2 Complete Anti-Federalist at 2.8.78 (cited in note 30)(letters from the "Federal Farmer")("In a federal system we must not only balance the parts of the same government, as that of the state, or that of the union; but we must find a balancing influence between the general and local governments."). 57 6 Annals of Cong. 1712-27 (Dec. 28, 1796). 18 Susan Rose-Ackerman has contended that local politicians in a decentralized system will be more risk averse than politicians in a consolidated national government. Susan Rose- Ackerman, Risk Taking and Reelection: Does Federalism Promote Innovation?, 9 J.Legal Stud. 593 (1980). The point here, however, is that there will be more innovation in a decentralized system as a whole, both because there are more actors and because individual con- stituencies will perceive risk and reward differently. This will hold true even if the average local politician is more risk averse than the average federal politician. 1987] Evaluating the Founders' Design 1499 payers, this means that there is a powerful incentive for decentralized governments to make things better for most people.59 In particular, the desire to attract taxpayers and jobs will promote policies of economic growth and expansion. It is well known, for example, that families often choose a community on the basis of the school system; a better school system encourages development and raises property values. Competi- tion among communities is therefore likely to result in superior education (as well as more cost-effective ways of providing it). This is especially likely given the strong business support for education. Because of the need for a well-educated work force, businesses often choose to locate in communities with a superior educational system and push for improved education in communities where they already have facilities." The chairman of Xerox Corporation has been quoted as saying, "Education is a bigger factor in produc- tivity growth [rates] than increased capital, economies of scale or '61 better allocation of resources. To be sure, the results of competition among states and locali- ties will not always be salutary. State-by-state determination of the laws of incorporation likely results in the most efficient forms of corporate organization6 2 but state-by-state determination of the law of products liability seems to have created a liability monster. This is because each state can benefit in-state plaintiffs by more generous liability rules, the costs being exported to largely out-of- state defendants; while no state can do much to protect its in-state manufacturers from suits by plaintiffs in the other states. Thus, competition among the states in this arena leads to one-sidedly pro-plaintiff rules of law. 3 The most important example of this phenomenon is the effect of state-by-state competition on welfare and other redistributive policies. In most cases, immigration of investment and of middle- to-upper income persons is perceived as desirable, while immigra-

" See Tiebout, 64 J.Pol.Econ. 416 (cited in note 37). '0 See Janice C. Simpson, A Shallow Labor Pool Spurs Businesses to Act to Bolster Education, Wall St.J. at 1, col. 1 (Sept. 28, 1987). 61 Id. 61 See Judge Ralph Winter, Private Goals and Competition Among State Legal Sys- tems, 6 Harv.J.L. & Pub.Pol. 127 (1982). '3 See Michael W. McConnell, A Choice of Law Approach to Products Liability Re- form, in Walter Olson, ed., New Directions in Liability Law (1988)(forthcoming). This conclusion hinges on several unstated assumptions, the most important of which is that manufacturers cannot (either because of legal restrictions or the possibility of arbitrage) set different prices in different states in response to product liability rules. See David A. Rice, Product Quality Laws and the Economics of Federalism, 65 B.U.L.Rev. 1, 5-8 (1985). 1500 The University of Chicago Law Review [54:1484

tion of persons dependent on public assistance is viewed as a drain on a community's finances. Yet generous welfare benefits paid by higher taxes will lead the rich to leave and the poor to come. This creates an incentive, other things being equal, against redistributive policies.6 4 Indeed, it can be shown that the level of redistribution in a decentralized system is likely to be lower even if there is virtually unanimous agreement among the citizens that higher levels would be desirable.6 5 Where redistribution is the objective, therefore, advocates should and do press for federal programs, or at least for minimum federal standards.66 Thus, the competition among states has an uncertain effect: often salutary but sometimes destructive. There are races to the bottom as well as races to the top. Often one's view of the allocation of authority for specific issues will depend on a prediction as to substantive outcomes rather than a general theory of federalism.

B. To protect "private rights" The most important reason offered by the defenders of state sovereignty was that state and local governments are better protec- tors of liberty. Patrick Henry went to the heart of the matter when he told the Virginia ratifying convention: You are not to inquire how your trade may be increased, nor how you are to become a great and powerful people, but how your liberties can be secured; for liberty ought to be the direct end of your Government.6 The most eloquent of the opponents of the Constitution, Henry stated flatly that in the "alarming transition, from a Confederacy to a consolidated Government," the "rights and privileges" of Americans were "endangered." 8 He was far from alone in this

4 To the extent that the poor migrate in response to job opportunities rather than welfare benefits, this effect is reduced. 6 Oates, Fiscal Federalism at 6-8 (cited in note 41); see also Paul E. Peterson, City Limits (1981); Posner, Economic Analysis of Law at 599, 611-12 (cited in note 42). This is an instance of the free rider problem: even if every member of the community would be willing to vote for higher welfare benefits, it would be in the interest of each to leave the burden of paying for the program to others. '6 This is not necessarily to say that welfare should be federalized. Against the point made in the text must be balanced the greater flexibility and humanity of programs con- ducted at the community level, especially through voluntary associations. An impersonal, bureaucratic welfare system poses grave risks to the character of both recipients and providers. I do not begin to have an answer to the problem. 67 Storing, 5 Complete Anti-Federalist at 5.16.2 (cited in note 30). 68 Id. 1987] Evaluating the Founders' Design fear.69 At a distance of 200 years, it is this aspect of the founders' thought that is most difficult for us to understand.7 0 After Brown v. Board of Education7' and the various civil rights acts, after the revolution in criminal procedure fostered by federal law and federal courts, after the imposition of uniform federal standards for basic liberties under the Bill of Rights, and after the proliferation of novel statutory "rights" arising from the interventions of the welfare-regulatory state, it is the federal government, not the states, that appears to be our system's primary protector of individual liberties. This seems to be the premise of the Fourteenth Amendment and of much of New Deal legislation.72 The view at the founding, however, was much more divided and ambivalent. Madison's most important contribution to the debate over ratification is his challenging argument that individual liberties, such as property rights and freedom of religion, are better protected at the national than the state level. The argument, presented princi- pally in Federalist10,1 3 is familiar to all, but is no less controversial for being familiar. It is one of the glaring lacunae in The Foun- ders' Design that Berger fails to explore the implications of this argument. It is especially odd, given Berger's autobiographical ob- servation that he had long associated "States' Rights" with "Southern condonation of lynchings, with official oppression of blacks, and with demagogues who duped their constituents" (p. 5). It was these nefarious features of state government that Madison intended to counteract through the Union. Madison's argument, greatly simplified, is that the most serious threat to individual liberty is the tyranny of a majority faction. Since any given faction is more likely to be concentrated in a particular locality, and to be no more than a small minority in the

" See, e.g., 1 Federal Convention at 340-41 (Luther Martin) (cited in note 54); Storing, 2 Complete Anti-federalist at 2.3.7 (Robert Yates and John Lansing, Jr.) (cited in note 30); id. at 2.9.22 (Brutus). Compare Tocqueville's analysis: "Local institutions are to liberty what primary schools are to science; they put it within the people's reach; they teach people to appreciate its peaceful enjoyment and accustom them to make use of it. Without local institutions a nation may give itself a free government, but it has not got the spirit of liberty." Tocqueville, Democracy in America at 63 (cited in note 33). 70 For a more detailed discussion, see Richard B. Stewart, Federalism and Rights, 19 Georgia L.Rev. 917 (1985). 71 347 U.S. 483 (1954). 7' See Cass R. Sunstein, Constitutionalism After the New Deal, 101 Harv.L.Rev. -, (1987)(forthcoming). 73 Madison put the argument forward earlier in an essay entitled "Notes on the Con- federacy," written in April, 1787. See 1 Letters and Other Writings of James Madison 325- 28 (1865). 1502 The University of Chicago Law Review [54:1484 nation as a whole, it follows that factional tyranny is more likely in the state legislatures than in the Congress of the United States. This argument is supplemented by others, based on the "proper structure of the Union" 74-deliberative representation, separation of powers, and checks and balances-that also suggest that the federal government is a superior protector of rights. Here I shall concentrate on the argument from the "extent . . . of the Union. '75 Madison's argument blunted the anti-federalists' appeal to state sovereignty as the guarantor of liberty. It was, however, only partially successful. Why? Modern public choice theory has cast some doubt on elements of Madison's theory. In particular, Madison's assumption that the possibility of minority tyranny is neutralized by majority vote re- quirements and that minority factions are inherently vulnerable to majority tyranny76 is undermined by studies showing that a small, cohesive faction intensely interested in a particular outcome can exercise disproportionate influence in the political arena.77 Madison underestimated both the dangers of minority rule and the defensive resources of minority groups. Moreover, some observers have suggested that the conditions of modern federal politics-especially the balkanized, issue-oriented conjunction of bureaucratic agencies and committee staffs-is especially susceptible to factional politics. Professor Richard Stewart has dubbed the result "Madison's Nightmare."7 8 Proponents of greater state sovereignty in 1787-89 may have been rightly skeptical of Madison's claims that there would be less danger of factional oppression at the federal level. But even taking Madison's fundamental insight as correct-and surely it has much to commend it-the argument on its own terms cautions against total centralization of authority in Washington. It points instead to a hybrid system in which states

7' Federalist 10 (Madison), in The Federalist Papers at 84 (cited in note 24). 7& Id. 76 Id. at 80: If a faction consists of less than a majority, relief is supplied by the republican principle, which enables the majority to defeat its sinister views by regular vote. It may clog the administration, it may convulse the society; but it will be unable to execute and mask its violence under the forms of the Constitution. When a majority is included in a faction, the form of popular government, on the other hand, enables it to sacrifice to its ruling passion or interest both the public good and the rights of other citizens. 7 See generally Mancur Olson Jr., The Logic of Collective Action (1965); James Q. Wilson, Political Organizations (1973); Bruce A. Ackerman, Beyond Carolene Products, 98 Harv.L.Rev. 713 (1985). 78 Stewart, 19 Georgia L.Rev. at 921 (cited in note 70). 19871 Evaluating the Founders' Design 1503 retain a major role in the protection of individual liberties. There are three basic reasons. 1. Liberty through mobility. Madison's argument demon- strates that factional oppression is more likely to occur in the smaller, more homogenous jurisdictions of individual states. But it does not deny that oppression at the federal level, when it occurs, is more dangerous. The lesser likelihood must be balanced against the greater magnitude of the danger. The main reason oppression at the federal level is more dangerous is that it is more difficult to escape. If a single state chose, for example, to prohibit divorce, couples seeking a divorce could move (or perhaps merely travel) to other states where their desires can be fulfilled. Oppressive measures at the state level are easier to avoid. Important recent examples of this phenomenon are the migration of homosexuals to cities like San Francisco, where they received official toleration, and the migration of individuals from Massachusetts to New Hampshire to escape high rates of taxation. A more contentious example is the regulation of abortion. If the power to regulate abortion is returned to the states, there is little likelihood of effective enforcement of anti-abortion laws, since permissive jurisdictions would attract business from more restrictive states. On the other hand, a nationwide rule-either voted by Congress or adopted by the courts as a construction of the due process clause-would have far more dramatic consequences. Recognition of this feature of decentralized decision making does not depend on any particular understanding of the substantive content of "liberty." For these purposes, liberty need not be equated with government inaction. "States' rights" does not imply minimalist government. Under a regime of decentralized decision making, it is more, not less, likely that communities will adopt a radical, controversial form of social organization. Santa Monica, California, for example, can adopt a form of socialism that is unlikely to command majority support in any state or the nation at large. To some, Santa Monica will be a beacon of (a particular form of) liberty; to others, it is a petty tyranny.79 Indianapolis can (or could, if the courts would allow it)8° adopt anti-pornography legislation more stringent than national norms. To some (a curious alliance of feminists and social conservatives) this protects their

7' The Santa Monica example is borrowed from Charles Fried, Federalism-Why Should We Care?, 6 Harv.J.L. & Pub.Pol. 1, 2 (1982). 80 American Booksellers Ass'n, Inc. v. Hudnut, 771 F.2d 323 (7th Cir. 1985), aff'd 106 S.Ct. 1172 (1986). 1504 The University of Chicago Law Review [54:1484 freedom from a pornography-ridden society; to others, this is a vio- lation of freedom of expression. The liberty that is protected by federalism is not the liberty of the apodictic solution, but the liberty that comes from diversity coupled with mobility. 2. Self-interested government. Madison held that there are two different and distinct dangers inherent in republican government: the "oppression of [the] . . . rulers" and the "injustice" of "one part of the society against . . . the other part.""' The first concern is that government officials will rule in their own interest instead of the interest of the people. The second is that some persons, organized in factions, will use the governmental powers to op- press others. Significantly, while Madison argued that the danger of faction is best met at the federal level (for the reasons summa- rized above), he conceded that the danger of self-interested representation is best tackled at the state level. "As in too small a sphere oppressive combinations may be too easily formed ag[ainst] the weaker party; so in too extensive a one, a defensive concert may be rendered too difficult against the oppression of those entrusted with the administration."82 Consequently, while powers most likely to be abused for factional advantage ought to be vested in the federal government, powers that are most likely to be abused by self-aggrandizing officials should be left in the states, where direct popular control is stronger. 3. Diffusion of power. Madison himself did not view his argument as establishing the superiority of a consolidated national government; rather, he presented his famous argument about the tyranny of factions in favor of the intermediate, federalist solution of dual sovereignty. In Federalist 51, he underscored that "the rights of the people" are best protected in a system in which "two distinct governments," federal and state, "will control each other."' 3 The diffusion of power, in and of itself, is protective of liberty. In Tocqueville's evocative words, "Municipal bodies and county ad- ministrations are like so many hidden reefs retarding or dividing

81 Federalist 51 (Madison), in The Federalist Papers 323 (cited in note 24); see also 1 Letters of James Madison at 325-28 (cited in note 75). 82 James Madison to Thomas Jefferson, Oct. 24, 1787, in 10 The Papers of James Madison 214 (Robert A Rutland, ed., 1977). For more extended analysis of this point, see Michael W. McConnell, Contract Rights and Property Rights: A Case Study in the Rela- tionship Between Individual Liberties and Constitutional Structure, - Calif.L.Rev. - (1988) (forthcoming); Andrzej Rapaczynski, From Sovereignty to Process: The Jurispru- dence of Federalism After Garcia, 1985 Sup.Ct.Rev. 341, 389. 83 Federalist 51 (Madison), in The Federalist Papers at 323 (cited in note 24). 1987] Evaluating the Founders' Design 1505 the flood of the popular will."' That the framers and ratifiers of the Constitution were not wholly persuaded that individual liberties are safer in the hands of the central government is evident from their provision of explicit protections for certain cherished liberties in the Bill of Rights. An instructive example is the freedom of religion. 5 If Madison's theory of factions is correct, it suggests that governmental authority over religion is more safely lodged in the federal government, where the multiplicity of religious sects will guarantee against oppression, than in the states, where a single religious denomination often enjoys majority support. Indeed, Madison used the example of religious sects to demonstrate his point in Federalist 10 and 51.86 The actual treatment of religious freedom in the Constitution is, however, diametrically opposed to the Madisonian model. State authority over religion was left intact. Madison proposed an amendment that "No State shall violate the equal rights of con- science, 87 even stating that this (along with speech, press, and jury trial rights against the states) was "the most valuable" of his proposed amendments to the Constitution.8 Notwithstanding his plea, the proposal was rejected by the Senate.89 By contrast, the federal government was forbidden to pass any law "respecting an establishment of religion"-that is, either establishing or disestab- lishing a religion-or prohibiting the "free exercise thereof."90 This was the "states' rights" approach to the religion question; it left decisions "respecting" the establishment of religion wholly to the states."e

0' Tocqueville, Democracy in America at 263 (cited in note 33). Compare Rapaczynski, 1985 Sup.Ct.Rev. at 389 (cited in note 82); Amar, 96 Yale L.J. at 1492-1519 (cited in note 8). 35 Though the debates over the religion clauses of the first amendment provide valuable insight into the framers' understanding of the connection between federalism and liberty, The Founders' Design makes no mention of them. 8' Federalist 10 (Madison), in The Federalist Papers at 84 (cited in note 24); Federalist 51 (Madison), in id. at 324. 87 1 Annals of Cong. 452 (June 8, 1789) (J. Gales, ed. 1834). Two printings exist of the first two volumes of the Annals of Congress. They contain different pagination, running heads, and back titles. The printing with the running head "History of Congress" conforms to the remaining volumes of the series while the printing with the running head "Gales & Seaton's history of debates in Congress" is unique. See Checklist of United States Public Documents 1789-1909, 1463 (3d ed. 1911). All page citations herein are to the latter printing. 88 Id. at 458. 89 1 Annals of Cong. 86 (Sept. 21, 1789). 90 U.S. Const. amend. I. 9' See Wilbur G. Katz, Religion and American Constitutions 8-11 (1964); William W. 1506 The University of Chicago Law Review [54:1484

This decision was understandable. While it was more likely that individual states would erect a religious establishment (indeed, at that time, five of the thirteen states had an establishment of some sort), a national establishment would have been far more threatening to religious liberty. Religious dissenters were free to travel to more tolerant states, and did; moreover, the example of the more tolerant states generated pressure on the more restrictive states to modify their policies. By 1834, the last state establishment was repealed. A national establishment would have been far more difficult to eradicate. Moreover, religious minorities are more likely to have influence in an individual state where they are concentrated, and thus more likely to have their rights respected, than at the national level. As "Philadelphiensis" said of those Quakers who feared the loss of their religious exemption from compulsory military service if control over the military were vested in Congress instead of the state legislature: "Their influence in the state of Pennsylvania is fully sufficient to save them from suffering very materially on this account; but in the great vortex of the whole continent it can have no weight." 2 The religious freedom example illustrates that, right or wrong, the framers of the Constitution and Bill of Rights believed that state governments were, in some vital respects, safer repositories of power over individual liberties than the federal government. It is thus no accident that the "police power"-the protection of public health, safety, and morals-was left to the states, with the federal government entrusted with less sensitive powers like those over interstate and foreign commerce. As Berger comments, "Moral issues ...are best left to the States, precisely as the Founders intended" (p.146). Given the diversity of views about issues of morality, and the potential for oppression, it is natural that lovers of liberty would be inclined toward decentralized decision making. At this point, an important qualification is in order. The arguments from the "public good" and from "private rights" make sense only if one presupposes that the decision in question is appropriate to democratic decision making at some level, be it state or federal. Some issues are so fundamental to basic justice that they must be taken out of majoritarian control altogether. This is why both state and federal governments are prohibited, for example, from passing ex post facto laws and bills of attainder.9 3 These

Van Alstyne, What Is "An Establishment of Religion?," 65 N.C.L.Rev. 909 (1987).

92 Storing, 3 Complete Anti-Federalist at 3.9.12 (cited in note 30). "I U.S. Const. art. I, § 9, cl.3; id., § 10, cl.1. 1987] Evaluating the Founders' Design 1507 issues are thus subject to a single national rule; the reason, however, has nothing to do with federalism. Federalism is a system for allocation of democratic decision making power. For those few but important matters on which democracy itself cannot be trusted, neither the "public good" nor the "private rights" argument for state autonomy can hold sway. Obviously, different people will assign wider or narrower lati- tude to majoritarian institutions. The alternative to democracy in our system is not utopia but judicial rule, which is not immune to abuse and which unavoidably conflicts with the ideals of republicanism, discussed below. The conclusion that states should retain a high degree of decision making autonomy is stronger on the hum- ble assumption that most governmental decisions are fairly debata- ble-that is, that there is no single compelling just answer to many questions of government. Even as to compelling matters of justice, however, federalism remains important as a tactical consideration, at least until a just national consensus emerges. Prior to a national majority against slavery, abolitionists would prefer state-by-state decision making, since there would be at least some free states. Upon emergence of an anti-slavery national majority, abolitionists would prefer national legislative power. Once a substantial national consensus de- veloped-manifested in two-thirds of both Houses of Congress and three-quarters of the states-it became time to take the issue out of democratic politics. But these judgments would not be princi- pled decisions about federalism; they would be tactical judgments about abolitionism. (On this analysis, the Constitution's allocation of power with respect to slavery was precisely what tactically- minded abolitionists should have wanted given the political circumstances. 4 I therefore believe critics of the framers' work as supportive of slavery are mistaken.95)

C. To Preserve "the Spirit and Form of Popular Government" It was an article of faith among advocates of state autonomy that republicanism could survive only in a small jurisdiction. As stated by the prominent anti-federalist essayist, "Brutus," "a free republic cannot succeed over a country of such immense extent,

" The Dred Scott decision, employing the pernicious doctrine of substantive due process to vitiate Congress's power to deal with the slavery question, upset the constitutional scheme and thus made civil war unavoidable. "I See Remarks of Thurgood Marshall at the Annual Seminar of the San Francisco Patent and Trademark Law Association (Maui, Hawaii, May 6, 1987). 1508 The University of Chicago Law Review [54:1484 containing such a number of inhabitants, and these increasing in such rapid progression as that of the whole United States."9 6 They believed consolidated national government would lead to aristo- cratic or despotic rule. Their reasons may be reduced to three major themes: (1) enforcement of laws, (2) nature of representation, and (3) cultivation of public spiritedness. 1. Enforcement of laws. Obedience to the law can arise from two different sources: fear of punishment and voluntary compli- ance. A republican government, which has a minimal coercive apparatus, must rely predominantly upon the latter. As Brutus explained, in a free republic "the government must rest for its support upon the confidence and respect which the people have for their government and laws."'97 To the advocates of decentralized government, this necessarily implied that the units of government must be small and close to the people. "The confidence which the people have in their rulers, in a free republic," according to Brutus, "arises from their knowing them, from their being responsible to them for their conduct, and from the power they have of displacing them when they misbehave."9 8 Unfortunately, this confidence is impossible in a country the size of the United States. The different parts of so extensive a country could not possi- bly be made acquainted with the conduct of their representatives, nor be informed of the reasons upon which measures were founded. The consequence will be, they will have no confidence in their legislature, suspect them of ambitious views, be jealous of every measure they adopt, and will not support the laws they pass.9 This proposition finds support in the folklore of the small town, which in contrast to the big city is an oasis of law abiding- ness and community good feeling. It also seems consistent with public choice theory, since in a smaller setting it is more likely that a strategy of cooperation will overcome the "prisoner's dilemma," which in this context holds that the optimal strategy for each citizen is to violate the law while all others abide by it. In a smaller jurisdiction, there is greater likelihood of monitoring and of stig- matization or retaliation, hence greater incentive to abide by legal

9' Storing, 2 Complete Anti-Federalist at 2.9.11 (cited in note 30). 9 Id. at 2.9.18. See also Gordon S. Wood, The Creation of the American Republic, 1776-1787, 66 (1969). 98 Storing, 2 Complete Anti-Federalist at 2.9.18 (cited in note 30). 99 Id. 1987] Evaluating the Founders' Design 1509 and other ethical norms. 100 2. Nature of representation. One of the principal arguments for substantial state autonomy was that representatives in a smaller unit of government will be closer to the people. Patrick Henry, for example, warned in the Virginia ratifying convention that "throwing the country into large districts . . . will destroy that connection that ought to subsist between the electors and the elected." 01 Assuming representative bodies of roughly the same number, any given representative will have fewer constituents and a smaller district at the state or local level. Each citizen's influence on his representative, therefore, will be proportionately greater, and geographically concentrated minorities are more likely to achieve representation. Because federal electoral districts must of necessity be larger and more populous, representation is likely to be skewed in favor of the well-known few-what were known at the time as the "aris- tocratic" element.102 The Federal Farmer argued that increasing the number of representatives would make the nation "more democratical and secure, strengthen the confidence of the people in it, and thereby render it more nervous and energetic.' 0 3 How- ever, the sheer size of the United States makes it impossible to increase the number of representatives sufficiently, without turning the Congress into what Madison called "the confusion of a 10 4 multitude. Moreover, if representatives to the national government are required to spend much of their time at the distant national capital, they are likely to lose touch with the sentiments of their constituents, and instead come to identify themselves with the interests of the central governmental apparatus. 05 Even Madison realized that "within a small sphere, this voice [of the people]

I'0 See Robert Axelrod, The Evolution of Cooperation (1984). 101 Storing, 5 Complete Anti-Federalist at 5.16.27 (cited in note 30). The Federal Farmer argued, similarly, that "a small representation can never be well informed as to the circumstances of the people, the members of it must be too far removed from the people, in general, to sympathize with them, and too few to communicate with them." Id. at 2.8.99. 102 See, e.g., Storing, 5 Complete Anti-Federalist at 5.6.2 (cited in note 30)(Patrick Henry); id. at 2.8.158 (the "Federal Farmer"); id. at 6.12.16 (Melancton Smith). 103 Id. at 2.8.158. ,0, Federalist 10 (Madison), in The Federalist Papers at 82 (cited in note 24). 101 Compare Storing, 2 Complete Anti-Federalist at 2.6.27 (cited in note 30)(Letters of Cato) with Tullock, Federalism: Problems of Scale at 25 (cited in note 38)("the longer the chain of officials that runs between the voter making the choice and the actual provision of the product, the more noise is introduced into the process by the individual bureaucrats who have their own preference functions and by the problems of information transmission"). 1510 The University of Chicago Law Review [54:1484 could be most easily collected, and the public affairs most accu- rately managed."'10 6 3. Public spiritedness. Critics of governmental centralization warned that public spiritedness-then called "public virtue" 0 7-- could be cultivated only in a republic of small dimensions. Republicanism, it was thought, depended to an extraordinary degree on the willingness of each citizen to submerge his own passions and interests for the common good.10 8 The only substitute for public virtue was an unacceptable degree of coercion, compatible only with nonrepublican forms of government. There were two reasons to believe that a centralized government would undermine republican virtue. First, public spiritedness is a product of participation in deliberation over the public good. If the citizens are actively engaged in the public debate they will have more of a stake in the community. The federal government is too distant and its compass too vast to permit extensive participation by ordinary citizens in its policy formulations. By necessity, decision making will be delegated to agents. But as they are cut off from active participation in the commonwealth, the citizens will become less attached to it and more inclined to attend to their private affairs. Second, the natural sentiment of benevolence, 109 which lies at the heart of public spiritedness, is weaker as the distance grows between the individual and the objects of benevolence. An individual is most likely to sacrifice his private interests for the good of his family, and then for his neighbors and, by extension, his community. He is unlikely to place great weight upon the well-being of strangers hundreds of miles away. It is unlikely, therefore, that citizens of a nation as large as the United States will assume an atti- tude of republican virtue toward national affairs. The Founders' Design is curiously oblivious to the debate between federalists and anti-federalists over the nature of republicanism. Although much of the ratification controversy was couched in terms of "republicanism" and "popular" government, and although this aspect of the debate powerfully supports Berger's substantive conclusions about the preservation of state autonomy, the

1 10 Papers of James Madison at 212 (cited in note 82). 107 See Wood, Creation of the American Republic at 68 (cited in note 97). 108 For the connection between this doctrine of public virtue and the framers' concep- tion of religious liberty, see Michael W. McConnell, Accommodation of Religion, 1985 Sup.Ct.Rev. 1, 14-22; Herbert J. Storing, What the Anti-Federalists Were For 22-23 (1981). 109 See Adam Smith, The Theory of Moral Sentiments (D.D. Rapheal, ed. 1976). 1987] Evaluating the Founders' Design 1511 book makes no reference to it whatsoever. This is symptomatic of Berger's anti-theoretical approach in The Founders' Design. We would learn much more if Berger would use his considerable histo- riographic skills to explain the philosophic basis for the founders' evident attachment to decentralized government.

IV. The argument for substantial state and local autonomy was powerful at the time of the founding, and remains so. Even though recent Supreme Court decisions mark the all-time low in respect for the constitutional principles of federalism, 110 there has been a revival of interest, across the political spectrum,"' in devolution of governing authority to state, city, and community levels. The Founders' Design thus appears at a propitious time. Consideration of the reasons for decentralized political decision making bolsters many of the conclusions of The Founders' Design, though a thorough analysis of federalism today would require, as well, a more systematic appraisal of the arguments for a centralized national authority. This I have not attempted. More- over, if The Founders' Design is to have any practical effect, as Berger implies is his intention (pp. 186-88), much more thinking needs to be done about the appropriate role of the judiciary, the Congress, and the states themselves. Berger's apparent vision that the Supreme Court, having been informed of the founders' intentions, now has it in its power to restore the original constitutional scheme, is fanciful, and would not necessarily be desirable even if it were less so. The Constitution is everyone's responsibility, and not just the Supreme Court's. Restoration of the constitutional order requires more than a history lesson directed to the Court. It requires a renewed sense by the people of the relation of state sovereignty to the public good, individual liberty, and popular government. Notwithstanding its weaknesses, The Founders' Design is a welcome addition to the bicentennial outpouring. The book pains-

,,o FERC v. Mississippi, 456 U.S. 742 (1982); Garcia v. San Antonio Metropolitan Transit Authority, 469 U.S. 528 (1985); South Dakota v. Dole, 107 S.Ct. 2793 (1987). See Mark Tushnet, Federalism and the Traditions of American Political Theory, 19 Georgia L.Rev. 981 (1985); A Symposium on Federalism, 6 Harv.J.L. & Pub.Pol. (special issue 1982); Richard A. Epstein, The Proper Scope of the Commerce Power, - Va.L.Rev. - - (1987)(forthcoming); Gerald E. Frug, The City as a Legal Concept, 93 Harv.L.Rev. 1059 (1980); Amar, 96 Yale L.J. 1425 (cited in note 8); Rapaczynski, 1985 Sup.Ct.Rev. 341 (cited in note 82); Stewart, 19 Georgia L.Rev. 917 (cited in note 70); Ostrom, The Political Theory of a Compound Republic (cited in note 31). 1512 The University of Chicago Law Review takingly collects the statements of the framers and ratifiers about each clause that contributes in a major way to the debate about the allocation of authority between the federal government and the states. Whatever our chosen theory of interpretation, it is good to cast our minds back to the time of the founding, when popular attention was directed, uniquely in our history, to the issues of self-government. It is the only way to recall, and perhaps recap- ture, what we may have lost.